Midterm IIsols

MIDTERM II: SOLUTIONS
MATH 3140
1. Let Z[ 3] = {a + b 3 : a, b Z}. Show that Z[ 3] is a ring under the ordinary

addition and multiplication of real numbers.
Solution. Z[ 3] is a subset of the ring (R, +, ). Let us first show that Z[ 3] is closed under
both + and . Indeed, we have
a + b 3 + a0 + b0 3 = (a + a0 ) + (b + b0 ) 3 Z[ 3]
and
(a + b 3) (a0 + b0 3) = (aa0 + 3bb0 ) + (ab0 + a0 b) 3 Z[ 3].
0
0
0
0
Moreover,
since
(a
+
b
3)
+
(a
b
3)
=
(a
a
)
+
(b
b
)
3
Z[
3], it follows that
(Z[ 3], +) is a subgroup of (R, +), and is thus an abelian group. (We are using the fact
that if G is a group, and S G is a subset, then S is a subgroup if and only if ab1 S for
all a, b S.)
To check that (Z[ 3], +, ) is a ring, we must check that (Z[ 3], +) is an abelian group
(which we have done above), that is associative (this is true since it is true for
R), and
that the distributive laws hold (this is also true since it is true for R). Thus (Z[ 3], +, ) is
a ring.

2. Factor x6 + 6 Z7 [x] into linear terms in Z7 [x].
Solution. Let f (x) = x6 + 6 Z7 [x]. By Fermats Theorem we have 6 1 (mod 7) for
all 0 6= Z7 . Thus f () = 0 for all 0 6= Z7 (note that this also follows easily by
inspection). It follows that (x ) divides f (x) for all 0 6= Z7 . Consequently
x6 + 6 = (x 1)(x 2)(x 3)(x 4)(x 5)(x 6)q(x) Z7 [x],
for some q(x) Z7 [x]. For reasons of degree, deg q(x) = 1. By considering the coefficient of
x6 , it is clear that q(x) = 1. Thus
x6 + 6 = (x 1)(x 2)(x 3)(x 4)(x 5)(x 6) Z7 [x].

3. Let F be a field and let K be a subset of F with at least two elements. Prove that K
is a subfield of F if for any a, b K with b 6= 0, then both a b and ab1 are in K.
Date: November 16, 2009.
1
Solution. To fix notation, we have the field F given by the collection (F, +, ). Consider the
subset K of the group (F, +). I claim that a b K for all a, b K. Indeed, this is true by
assumption unless b = 0, in which case a b = a K. It follows that (K, +) is a subgroup
of (F, +), and hence is an abelian group.
Now consider K = K {0} and F = F {0}. We know that (F , ) is a group. By
assumption K is a non-empty subset of this group with the property that ab1 K for all
a, b K . In fact, since F is an integral domain, it must be that ab1 K for all a, b K .
Thus (K , ) is a subgroup of (F , ).
It is also true that K is closed under the operation . Indeed, since K is closed under ,
it remains only to observe that a 0 = 0 a = 0 K for all a K (recall that 0 K since
(K, +) (F, +)).
To check that (K, +, ) is a ring, we must check that (K, +) is an abelian group (which
we have done in the first paragraph), that is associative (this is true since it is true for F ),
and the distributive laws hold (this is also true since it is true for F ). Thus K is a subring
of F . It follows that K is a commutative ring.
Now since K is a subgroup of F it contains the multiplicative identity 1 6= 0 and every
element a K has a multiplicative inverse a1 K . Thus K is a subfield of F .

4. True or false. If true, prove the statement. If false, provide a counter example.
(a) If d||G| then there exists a g G such that |g| = d.
(b) Suppose R is a ring and a, b R. If ab = 0 then either a = 0 or b = 0.
Solution. (a) and (b) are both false. For (a) consider the group G = Z2 Z2 . Then the
number 4 divides |G| = 4. On the other hand, every element of G has order at most two.
For (b) consider the ring Z4 . We have [2][2] = [4] = [0] Z4 , and [2] 6= [4].

5. Let G be a group. Show that if G/Z(G) is cyclic, then G is abelian.
Proof. To show G is abelian, we must show that given g1 , g2 G, then
g1 g2 = g2 g1 .
To begin, since the group G/Z(G) is cyclic, it has a generator [g] G/Z(G) for some g G.
It follows that there are integers n1 , n2 such that
[g1 ] = [g]n1 and [g2 ] = [g]n2 .
We can rewrite this by saying that there exists z1 , z2 Z(G) such that g1 = g n1 z1 and
g2 = g n2 z2 . Then
g1 g2 = g n1 z1 g n2 z2 = g n2 z2 g n1 z1 = g2 g1
since by definition z1 , z2 commute with all elements of G, and g commutes with itself.

6. An element of a of a ring R is nilpotent if an = 0 for some n N. Show that if a R
is nilpotent, then 1 a has a multiplicative inverse in R.
Solution. Using the condition an = 0, we have
(1 a)(1 + a + a2 + . . . + an1 ) = 1 an = 1.
2
Thus 1 + a + a2 + . . . + an1 is the multiplicative inverse of (1 a).
7. Show that An is a simple group for n 5.

Solution. We break this problem into several parts.
Claim (a): An contains every 3-cycle if n 3.
Proof. Let (a1 , a2 , a3 ) Sn be a 3-cycle. Since (a1 , a2 , a3 ) = (a1 , a2 )(a3 , a2 ) it follows from
the definition that (a1 , a2 , a3 ) An .

Claim (b): An is generated by the 3-cycles.
Proof. Let An be a nontrivial element. By definition there is an expression of
= 1 2 2n1 2n
as a composition of transpositions 1 , . . . , 2n for some n N. Since there are n-pairs of transpositions in the expression, the claim will follow if we can show that for any transpositions
, Sn with 6= , then is a composition of 3-cycles.
To prove this, suppose = (a1 , a2 ) and = (a3 , a4 ). There are two cases to consider:
(1) If ai 6= aj for i, j {1, 2, 3, 4} and i 6= j, then (a1 , a2 )(a3 , a4 ) = (a1 , a3 , a2 )(a1 , a3 , a4 ).
(2) Otherwise ai = aj for some i 6= j, and we can assume without loss of generality that
a2 = a4 . Then we have (a1 , a2 )(a3 , a2 ) = (a1 , a2 , a3 ).
Thus is a composition of 3-cycles, completing the proof of Claim (b).

Claim (c): Fix r, s {1, . . . , n} with r 6= s. If n 3, then An is generated by the set of
3-cycles {(r, s, i) : 1 i n}.
Proof. After some manipulation, one can establish the identities:
(i) (r, s, i)2 = (s, r, i),
(ii) (r, s, j)(r, s, i)2 = (r, i, j),
(iii) (r, s, j)2 (r, s, i) = (s, i, j),
(iv) (r, s, i)2 (r, s, k)(r, s, j)2 (r, s, i) = (i, j, k).
Since every 3-cycle is of the form of one of those above, it follows that An is generated by
the set of 3-cycles {(r, s, i) : 1 i n}.

Claim (d): Suppose n 3. Let N C An be a normal subgroup. If N contains a 3-cycle
then N = An .
Proof. Suppose N contains a 3-cycle . Then = (r, s, i) for some choice of r, s, i
{1, . . . , n}. Observe (after some manipulation) that for any j 6= i {1, . . . , n} we have
((r, s)(i, j)) (r, s, i)2 ((r, s)(i, j))1 = (r, s, j).
The expression on the left in in N since it is a conjugate of an element of N . Thus N
contains the set {(r, s, j) : 1 j n}. By virtue of Claim (c), it follows that N = An .
Claim (e): Suppose n 5. If N C An is a non-trivial normal subgroup, then N contains a
3-cycle.
3
Proof. We will do this in a case by case analysis. The first step is to show that if N C An is
a non-trivial normal subgroup, then one of the following cases holds:
CASE I: There exists N that can be written as a disjoint product of the form =
(a1 , . . . , ar ) for some r 4.
CASE II: There exists N that can be written as a disjoint product of the form
= (a4 , a5 , a6 )(a1 , a2 , a3 ).
CASE III: There exists N that can be written as a disjoint product of the form
= (a1 , a2 , a3 ), with a disjoint product of transpositions.
CASE IV: There exists N that can be written as a disjoint product of the form
= (a3 , a4 )(a1 , a2 ), with a disjoint product of transpositions.
To see that one of these cases must hold, consider the fact that any non-trivial Sn can
be written as a product of disjoint cycles
= 1 . . . m
for some m N. Since disjoint cycles commute, we may reorder so that the length of the
cycles is non-decreasing. The fact that one of the cases above must hold is then obvious.
Now we will show that in each case above, N contains a 3-cycle. For Case I, consider
the expression 1 (a1 , a2 , a3 )(a1 , a2 , a3 )1 . This is in N since (a1 , a2 , a3 )(a1 , a2 , a3 )1 is a
conjugate of an element of N . On the other hand, after some algebra, one has
1 (a1 , a2 , a3 )(a1 , a2 , a3 )1 = (a1 , a3 , ar ),
so that N contains a 3-cycle.
For Case II, consider the expression 1 (a1 , a2 , a4 )(a1 , a2 , a4 )1 . Again this is clearly in
N , and after some algebra one has
1 (a1 , a2 , a4 )(a1 , a2 , a4 )1 = (a1 , a4 , a2 , a6 , a3 ).
Thus N contains a cycle of length five, and so by Case I, it also contains a cycle of length
three.
For Case III, one has
2 = (a1 , a3 , a2 )
using the fact that 2 is the identity (it is the product of disjoint transpositions). Thus N
contains a 3-cycle.
Finally, for Case IV, consider 1 (a1 , a2 , a3 )(a1 , a2 , a3 )1 . Some algebra shows that this
is equal to (a1 , a3 )(a2 , a4 ). We call this permutation , which as above, is in N . Now let
= (a1 , a3 , i) for some i {1, . . . , n} {a1 , . . . , an }. Then
1 = (a1 , a3 , i),
which again is in N for the same reason. Thus N contains a 3-cycle.
Let us conclude by showing that An is simple for n 5. Let N C An be a non-trivial

normal subgroup of An . In Claim (e) we showed that such a subgroup must contain a 3-cycle.
In Claim (d) we showed that if N contains a 3-cycle, then it is equal to An . This proves that
the only normal subgroups of An are the trivial subgroup and An . Thus An is simple.

Midterm IIsols

Uploaded by

Copyright:

Available Formats

Midterm IIsols

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Midterm IIsols

Uploaded by

Copyright:

Available Formats

MIDTERM II: SOLUTIONS

1. Let Z[ 3] = {a + b 3 : a, b Z}. Show that Z[ 3] is a ring under the ordinary

(a + b 3) (a0 + b0 3) = (aa0 + 3bb0 ) + (ab0 + a0 b) 3 Z[ 3].

Thus 1 + a + a2 + . . . + an1 is the multiplicative inverse of (1 a).

7. Show that An is a simple group for n 5.

Let us conclude by showing that An is simple for n 5. Let N C An be a non-trivial

You might also like